Single-file relationships vs. Multi-file: SPEED

[JamesL]

In version 5.5 I made a multifile relational Database according to the FMP example files "Inventory", Items", and "Sales". Now in the "Inventory" file I do a find request in the unstored calculation field "Quantity on Hand" (Initial Inventory Level - Quantity Sold).

I notice that the more records the "Inventory" database contains, the longer it takes to find the records that show the quantity on hand of my find request. In fact, it took too long, so in FMP 5.5 I abandoned the idea of using it.

Now I just started studying the trial version of FMP 7, my question before I upgrade: As to the SPEED of find requests on unstored calculation fields, will this greatly improve if the 3 seperate example files are placed into tables of a single database (FMP 7)

[Ender]

It should not matter if the tables are in one file or many, finds on unstored calcs should perform equally well (much faster than in FM5/6.) What can improve speeds of such searches is using FileMaker Server 7.

HarryGlos posted his real-world comparisons here , where he concludes that separate files perform better for large solutions.

[Vaughan]

That HarryGlos is quite a character...

[JamesL]

Thank you so much Ender, for your good advise, as well as for the link to the HarryGlos thread, quite funny and most interesting.